[feature request] alternative compression methods for backups
- Thread starter pille
- Start date
of course:
different compression algorithms implement different trade-off between size and time/load.
i want to be able to choose the solution that's best for me.
take an ubuntu kvm backup (tar=1.9G):
486M lzma (default options, 18min)
634M bz2 (-9, 5min)
695M gz (your default options, 1.5min)
for me it's ok to spend more time for -200M per backup on multi-cpu-machines.
and: it seems not that hard to implement, so it's worth the effort.
different compression algorithms implement different trade-off between size and time/load.
i want to be able to choose the solution that's best for me.
take an ubuntu kvm backup (tar=1.9G):
486M lzma (default options, 18min)
634M bz2 (-9, 5min)
695M gz (your default options, 1.5min)
for me it's ok to spend more time for -200M per backup on multi-cpu-machines.
and: it seems not that hard to implement, so it's worth the effort.
Last edited:
no, if i had one, i would have posted it. sorry for the misunderstanding.
in the backend just exchange gzip with bzip2 or lzma.
the frontend is a bit more complicated, but that's where you are experts.
either implement a dropdown box to select a method or give the user a field to enter a compression command (including options) in the pipe.
in the backend just exchange gzip with bzip2 or lzma.
the frontend is a bit more complicated, but that's where you are experts.
either implement a dropdown box to select a method or give the user a field to enter a compression command (including options) in the pipe.
its from minor priority here, so you can either send a patch, pay someone doing it or wait til we got time to do it. I think storage space is cheaper than CPU and I/O.